If your term has a 20-mark periodic test, a 10-mark quiz and an 80-mark half-yearly in it, you cannot just add the marks up and you cannot just average the percentages. Both of those quietly decide how much each test counts, and neither of them asks you what you intended.
The method that works is two steps: convert every test to a fraction of its own maximum, then multiply each fraction by the weight you meant it to carry. This article works through one student's term with real numbers, shows that the three common methods land 9 percentage points apart, and explains exactly where the gap comes from.
One student, five assessments
Here is Aarav, Class 9 Science, one term. Five assessments, five different maximum marks - which is completely normal and is where the trouble starts.
| Assessment | Maximum | Obtained | Score |
|---|---|---|---|
| Periodic Test 1 | 20 | 18 | 90.0% |
| Periodic Test 2 | 25 | 15 | 60.0% |
| Class quiz (online) | 10 | 10 | 100.0% |
| Practical | 30 | 21 | 70.0% |
| Half-yearly | 80 | 49 | 61.3% |
| Total | 165 | 113 | - |
Three methods, three different answers
Ask three teachers for Aarav's term percentage and you will plausibly get three numbers.
Method A - average the percentages
Convert each test to a percentage, then take the mean of the five.
Method B - total obtained over total maximum
Add up everything he scored, divide by everything that was available.
Method C - weight each assessment deliberately
Decide first what each assessment is worth, then apply it. Say the school intends periodic tests to carry 10% each, the quiz 5%, the practical 15% and the half-yearly 60%.
Same student, same marks, same term: 76.3%, 68.5% and 67.3%. In most grading scales that spread crosses at least one grade boundary. This is not a rounding technicality - it is a nine-point decision, and if nobody wrote it down then it was made by accident.
Where the nine points come from
Every method applies weights. Methods A and B just do not tell you what they are. Here is what each one actually did to Aarav:
| Assessment | You intended | Method A applied | Method B applied |
|---|---|---|---|
| Periodic Test 1 | 10% | 20% | 12.1% |
| Periodic Test 2 | 10% | 20% | 15.2% |
| Class quiz (online) | 5% | 20% | 6.1% |
| Practical | 15% | 20% | 18.2% |
| Half-yearly | 60% | 20% | 48.5% |
Method A gives every assessment an equal share. Method B gives each one a share equal to its maximum marks divided by 165.
Method A treats a 10-mark quiz as equal in importance to the half-yearly. That is why Aarav's number jumps: he scored 10 out of 10 on the smallest thing in the term, and averaging percentages hands that a full 20% of his grade.
Method B is closer, but it is only right by coincidence. It weights the half-yearly at 48.5% because 80 is 48.5% of 165 - not because anyone decided the half-yearly was worth about half the term. If you ever change a paper from 80 marks to 100, Method B silently re-weights your entire term.
Method B is genuinely correct in exactly one case: when you set the maximum marks specifically so that they equal the weights you wanted. Some schools do design their assessments that way, and if yours does, Method B and Method C give the same answer. The point is to know which situation you are in.
The method to use
where the intended weights add up to 100
In practice:
- Write the weights down before the term starts. This is the actual decision. Everything after it is arithmetic.
- Check they sum to 100. If they do not, nothing downstream will be interpretable.
- Convert each assessment to a fraction of its own maximum - 18 out of 20 becomes 0.900.
- Multiply each fraction by its weight and add them up. The result is already a percentage.
The reason this is worth the extra column is that it separates two things that Methods A and B fuse together: how hard a paper was to mark out of, and how much it should matter. A quiz can be out of 10 and still be worth 5% of the term. A practical can be out of 30 and be worth 15%. Once the weights are explicit, you can change a paper's total marks without changing anyone's grade.
Three things that will come up
A student misses an assessment
Do not enter a zero unless you actually mean the student earned nothing. Instead, drop that assessment and rescale the remaining weights.
Suppose Aarav misses the practical. The remaining weights are 10 + 10 + 5 + 60 = 85, and he earned 9.00 + 6.00 + 5.00 + 36.75 = 56.75 of them.
That is the honest answer: it reports how he did on what he actually sat. Entering a zero for the practical would have given him 56.8% and described a student who does not exist. This is the difference between absent and scored nothing, and it is worth recording them as two different things rather than one, because you cannot recover the distinction later.
Some tests were on paper and some were online
The arithmetic does not care. A 30-mark practical marked by hand and a 10-mark online quiz enter the same formula in the same way - fraction of maximum, times weight. What matters is that both are in the same record by the time you calculate, because a term percentage assembled from two systems is a term percentage nobody can check. Keeping offline and online results in one place is worth more here than any individual feature.
Rounding
Round once, at the end. If you round the half-yearly to 61% before weighting it, Aarav's term comes out at 67.1% instead of 67.3%. That particular gap is small and will not change a grade on its own - but it costs nothing to avoid, and the same rounding repeated across six subjects and then again at the report level is how two people arrive at two different totals from the same mark sheet.
Decide the weights, then publish them
Almost every dispute about a term percentage is really a dispute about weights that were never agreed. The calculation takes a minute. The disagreement about whether the quiz should have counted for a fifth of the term takes a parent meeting.
So write the weighting down at the start of the term, put it where teachers and students can see it, and use the same one for every student in the class. If you are reviewing how your assessments are set up more broadly, our guide to best practices for conducting online school assessments covers the surrounding process, and the features a principal should look for covers what to expect a system to handle for you.
Teemly applies weights this way across both online tests and marks entered from paper exams, so a term percentage comes out of one record rather than being reassembled by hand each term. You can see how that works on our assessment software for schools page.